self-poisoning during pregnancy as a model for teratogenic ... · pdf filesex: female excess...
TRANSCRIPT
Self-poisoning during pregnancy
as a model for teratogenic risk estimation of
drugs
Disaster epidemiology
Prof. Andrew E. Czeizel, M.D., Ph.D., Doct. Sci.
Budapest, Hungary
Training course in sexual and reproductive health research
Geneva 2011
Dilemmas at the teratogenic risk estimation of drugs
1. Clinical trial programs of a drug under development cannot
include pregnant women.
2. Results of experimental animal investigations cannot be
extrapolated for pregnant women due to species differences.
Conclusion: the harsh reality is that human beings remain the
ultimate test organism to detect teratogenic/fetotoxic drugs.
Dilemmas at the teratogenic risk estimation of drugs
Postmarketing evaluation
3. Clinical observations /trials
case report – selection bias
clinical series – no appropriate control
RCT – no pregnant women 4. Epidemiological studies
Descriptive – not appropriate for identification of causes
Analytical – recall bias
small (clinical) doses
confounders Conclusion: these postmarketing data are useful to predict human teratogenic risk but the extent of these predictions needs to be taken with caution in the clinical practice.
Disaster epidemiology
a) Natural catastrophes/circumstances
earthquake
high radon exposure
high altitude, etc.
b) Human-made catastrophes/circumstances
nuclear bomb (Hiroshima, Nagasaki)
nuclear plant accident (Chernobyl)
extreme occupational exposures
self-poisoning (suicide)
Suicide
Suicide attempt
age: young (17-19 yr)
sex: female excess (1 : 4)
psychiatric diseases: rare
Suicide mortality
advanced age (over 50 yr)
male excess (2.5 : 1)
frequent
Budapest Monitoring of Self-
Poisoned Pregnant Women
(1960-1993)
Department of Toxicological Internal Medicine, Korányi Hospital, Budapest (”Study hospital”).
“Catchment” region: Budapest and surrounding area (Pest county): 3 million people.
In general about 100 pregnant women attempted suicide in each year during the study period.
Budapest Monitoring of Self-Poisoned Pregnant Women
(1960-1993)
Method:
1. Sensitive blood pregnancy test in all women aged between 15 and 50 years
2. A signature for informed consent
3. Special antenatal care
4. Data collection in a personal card Personal data. Lifestyle (alcohol, smoking, narcotics). Case history; History of previous pregnancies. History of study pregnancies; Self-poisoning: date and time (hour, minute); postconceptional fetal age; name, dose and administration route of chemicals (medicines); interval between intake of drugs and admission to the study hospital; antidotes.
Budapest Monitoring of Self-Poisoned Pregnant Women
(1960-1993)
5. Clinical evaluation of mothers Symptoms (e.g. duration of unconsciousness) Blood level of drugs Disease-process Pregnancy outcome (termination of pregnancy, miscarriage)
6. Follow-up of exposed children Birth outcomes (sex, gestational age, birth weight) Congenital anomalies, particularly congenital abnormalities (CAs) Postnatal development School records Psychometric examination Behavioral examination
7. Controls - a) Sibs b) Matched controls
Estimation of severity of self-poisoning
1. Blood concentration of drugs
however, it depends on time elapsed between self-poisoning and hospital admission, the effect of medical intervention in the ambulance car, treatment after admission to the study hospital (gastric lavage, antidotes, etc.).
2. Information of mothers regarding the number of drug tablets used for self-poisoning.
3. Clinical symptoms.
Classification of severity of self-poisoning
Mild no comatose condition at or after admission
Moderate comatose condition or unconsciousness at or after admission
Severe unconsciousness longer than 1 day after admission and/or artificial respiration
Very severe life-threatening: unconsciousness more than 2 days and/or uremia or multiorgan failure
Fatal
Total data set (1960-1993) No. of self-poisoned pregnant women
1,044
Died
19 (1.8%) 1960-79 (6.7%) 1980-84 (1.1%) 1985-93 (0.4%)
False address
99 (9.5%)
Evaluated pregnancies
926 (88.7%)
Livebirth
411 (44.4%)
Fetal death
162 (17.5%)
Termination of
pregnancy
353 (38.1%)
Adoption
41 (10%)*
Refused
collaboration
11 (2.7%)
Died
10 (2.4%)
Evaluated
365 (88.8%)
*16 were evaluated due to the data of pediatric status
Variables Self-poisoned pregnant women
Hungarian pregnant women
Quantitative Mean S.D. Mean S.D.
Maternal age 23.1 5.7 25.4 4.9
(19 year or less, %
32.1 8.6
Birth order 1.4 0.9 1.7 0.9
Categorical % %
Unmarried 58.0 4.9
Socioeconomic status
High 19.7 38.0
Medium 37.5 30.6
Low 42.8 31.4
Smokers 42.0 18.9
Regular drinkers 15.2 0.6
Characteristics of case mothers
Time of self-poisoning, in addition number of fetal losses
and livebirths (between 1960 – 1993)
Fetal
development
(lunar month)
Self-poisoning Fetal loss
Livebirt
h
No.
No. % Chemical
pregnancy
No.
Miscarriag
e No.
Induced
abortion
No.
I. 213 37.9 111 3 73 12
II. 128 22.8 3 6 94 15
III. 71 12.6 0 5 30 28
IV. 40 7.1 0 4 5 26
V. 30 5.3 0 0 1 27
VI. 28 5.0 0 0 0 21*
VII. 23 4.1 0 0 0 18*
VIII. 22 3.9 0 0 0 4*
IX-X. 7 0.4 0 0 0 2
Total 562 100.0 114 18 203 178
Distribution of drugs used for self-poisoning Drugs No. of liveborn babies Diazepam 112
Phenobarbitals 40
Nitrazepam 36
Meprobamate 27
Chlordiazepoxide 25
Promethazine 23
Valeriana + phenobarbital 20
Promethazine + glutethimide + amobarbital 15
Glutethimide 10
Amobarbital 7
Aminophenazone + phenacetin + caffeine 4
Medazepam 3
Barbital + aminophenazone 3
Others (in two or one pregnant woman) 46
Total 391*
*several pregnant women used more than one drug
Controversial estimation of teratogenic
potential of diazepam
• U.S. and Finnish studies in 1975 showed an association between
diazepam and orofacial clefts (cleft lip + palate and posterior cleft
palate).
• Laegreid et al. (1987-1992) delineated a benzodiazepine
embryofetopathy (with a cardinal symptom of orofacial clefts).
• In Hungary we were not able to detect any teratogenic effect of diazepam
and other benzodiazepines after the use of the usual clinical doses.
• Some other studies (e.g. in the USA) resulted only in negative findings.
Time and severity of self-poisoning with diazepam,
number of livebirths and congenital abnormalities
Fetal development (lunar month)
Livebirth No.
Mild-moderate
Severe- very severe
Congenital abnormalities
I. 9 1 8 1
II. 11 5 6 4
III. 18 2 6 0
IV. 12 7 5 2
V. 15 9 16 1
VI. 16 8 8 2
VII. 13 7 5 3
VIII. 11 1 17 2
IX-X. 7 0 7 0
Total 112 40 72 15
% 100.0 35.7 64.3 13.4
Time of self-poisoning
Critical period of
given CAs
5
After critical period
of CAs
10
Evaluation of 15 exposed children with CAs
Evaluation of 15 exposed children with CAs
Mild Severe FAS (fetal alcohol syndrome)
Unrecognised syndrome
Exposed 10 1 2 2
cong.inguinal hernia 3 undescended testis 2 talipes equinovarus 2 cong.dislocation of hip 1 ventricular septal defect (spontaneous closure) 2 hypospadias, glandular 1
pyloric stenosis Talipes equinovarus, lowbroad nasal bridge, gingivobuccal frenulum rectus diastasis Torticollis, high-arched palate, gingivobuccal frenulum, rectus diastasis
Controls 6 2 0 1
cong.inguinal hernia 2 undescended testis torticollis talipes equinovarus syndactyly
transposition of great vessels cong.arthrogryposis
(cleft lip + palate, complex cardiovascular CA, hypospadias)
Conclusion: no teratogenic effect (e.g. orofacial clefts)
Isolated CAs Multiple CAs
Severity
Controversial estimation of teratogenic
potential of diazepam
• U.S. and Finnish studies in 1975 showed an association between diazepam
and orofacial clefts (cleft lip + palate and posterior cleft palate).
• Laegreid et al. (1987-1992) delineated a benzodiazepine embryofetopathy
(with a symptom of orofacial clefts).
• In Hungary we were not able to detect any teratogenic effect of diazepam
and other benzodiazepines after the use of the usual clinical dates.
• Some other studies (e.g. in the USA) resulted only in negative findings.
• Now we can show that 38 pregnant women who attempted suicide with
very high dose of diazepam during the critical period of orofacial clefts had
no liveborn babies with these CAs.
Possible explanation
Diazepam was used in psychiatric patients in the
mentioned U.S. and Scandinavian studies while
diazepam was used for the treatment of threatened
abortion of pregnant women without any
psychiatric diseases in Hungary.
Confirmation
Our recent study showed that panic disorder can induce CAs
(mainly cleft lip ± palate) which can be prevented by diazepam.
Diazepam
– +
Panic Referent 1.1 (0.7 – 1.5)*
disorder + 3.1 (1.4 – 6.9)* 1.5 (0.7 – 3.2)*
*adjusted OR with 95% CI
Conclusion
The previously found teratogenic effect of diazepam
may be associated with underlying maternal diseases,
lifestyle and/or other confounders.
Components of Tardyl®,
a frequently used medicinal product for
insomnia
Tardyl® Dose
Amobarbital 125 mg
Glutethimide 125 mg
Promethazine 7.5 mg
Large doses of these components had no teratogenic
effect
Main findings of exposed children born to pregnant women who attempted
suicide with Tardyl® during pregnancy and of their exposed sibs
Exposed children
(N=27)
Unexposed sibs
(N=46)
Comparison
Categorical No. % No. % OR 95% CI
Congenital
abnormalities
2 7.4 2 4.4 1.8 0.2 – 13.9
Mental retardation 8 29.6 0 0.0 27.6 3.3-232.4
Quantitative Mean S.D. Mean S.D. p =
Birth weight (g) 2,883 55 2,895 67 0.94
Pregnancy age (wk) 36.4 2.8 36.8 2.9 0.79
Estimated
Cognitive status (IQ) AM M UM MR AM M UM MR p=
No. 3 7 4 8 4 12 4 0 0.38
IQ (mean + S.D.) 82.2 + 20.0 100.0 + 9.7 0.04
Behavioral scale N + ++ +++ N + ++ +++ p=
No. 4 4 3 5* 9 6 1 0 0.03
*5 mentally retarded children
Pregnancy history of mothers who attempted
suicide by Tardyl® in two pregnancies
1940 1949
1982
1974
1970
1968
1964
1958
3,350 g
40 wk
IQ:”Normal”
1958
3,350 g
40 wk
IQ:”Normal”
1977
2,000 g
34 wk
IQ: 96
1975
2,550 g
36 wk
IQ: 102
1972
1,000 g
30 wk
IQ: 108
1967
2,000 g
36 wk
IQ: 55
Undescended
testis
1965
2,650 g
37 wk
IQ: 65
Criminal
behaviour
Suicide
attempt in
20th gest. week
Suicide
attempt in
20th gest. week
Conclusions
Large doses of Tardyl® associated with a high risk (OR with 95%
CI: 27.6, 3.3-232.4) of mental retardation.
This association shows the importance of drug interactions (e.g.
glutethimide) because the components of Tardyl® separately
did not induce mental retardation.
Tardyl® is the first known mental retardation induced drug.
Strengths of self-poisoning model
1. Large doses are administrated once on a certain day of pregnancy.
2. These pregnant women are hospitalized and medical data are
available.
3. We can estimate the effective dose of drugs used for self-poisoning.
4. We can use the well-known dose-effect correlation: if very high
doses of a given drug during the critical period do not induce CA –
it is an important argument against the teratogenic potential of the
drug.
5. We can evaluate other variables of reproductive toxicology
(intrauterine growth retardation, mental and behavioral
development, chromosomal aberrations, second generation).
Limitations of self-poisoning model
1. High proportion of early fetal loss.
2. Many pregnancies are terminated (but the recent use of
abortion pills may help evaluate embryos).
3. Thus the number of liveborn infants born to mother who
attempted suicide during the critical period of CAs is
limited.
4. In general more than one drug is used for self-poisoning.
5. Several pregnant women who attempt suicide smoke
cigarettes, consume alcohol, etc.
Recommendations to establish an international monitoring system of self-
poisoned pregnant women
1. It is feasible (self-poisoned persons are cared in specialized inpatient clinics).
2. Self-poisoned patients are thoroughly examined and we can estimate the exposure doses of drugs.
3. Self-poisoned pregnant women are at high risk, therefore they need special medical care.
4. This model would be the most effective epidemiological approach to evaluate the teratogenic and mutagenic effect of drugs in human beings.
5. This international monitoring system may stimulate and contribute the research and prevention of suicidal behaviours.